Med-Term: Synthetic Multilingual Medical Notes for ASR Benchmarking

Med-Term is a fully synthetic evaluation dataset for automatic speech recognition released by Corti ApS. It contains medical notes dictated via text-to-speech in German, French, and English, with no real patient data or PHI. The dataset was released alongside the Symphony for Speech Recognition white-paper.

Use Cases

• Benchmark ASR system performance based on synthetic medical audio in three languages.
• Evaluate NLP model robustness on transcribed medical notes generated via TTS.
• Test multilingual speech processing pipelines on domain-specific synthetic content.

Strengths

• Contains synthetic data in three languages: German, French, and English.
• Explicitly contains no real patient data or personally identifiable information (PHI).
• Released alongside a named white-paper (Symphony for Speech Recognition).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license information are unknown.
• Data is fully synthetic, which may not capture all nuances of real-world medical speech.

Provenance

Source

Corti ApS

Collection Method

Fully synthetic generation using text-to-speech.

Freshness

Last updated 2026-06-10 07:45:21.